Superintelligence
《超级智能》
Writing before superintelligence existed, Bostrom systematically asked how much leverage humanity would retain once an intelligence far beyond our…
Artificial Intelligence
Classic Index 94
AGI
通用人工智能
Beyond-human general intelligence: possibility, risk, and the control problem.
4 books
Superintelligence
《超级智能》
Writing before superintelligence existed, Bostrom systematically asked how much leverage humanity would retain once an intelligence far beyond our…
Artificial Intelligence
Classic Index 94
Human Compatible
《AI 新生》
Russell proposes a solution that looks like a concession but is in fact demanding: do not give machines explicit objectives; make them uncertain…
Artificial Intelligence
Classic Index 90
Life 3.0
《生命 3.0》
Tegmark organizes a very broad question — from near-term automation and unemployment to a cosmos-scale intelligence — with a simple frame: life 1.0…
Artificial Intelligence
Classic Index 88
The Alignment Problem
《人机对齐》
Christian splits alignment into three layers: the objectives a model learns may not be the ones we intend, the biases it inherits from data may…
Artificial Intelligence
Classic Index 87